Before I picked up disc, I was a dps main. Playing “optimally” as a dps was essentially based on two things: swapping to priority targets (and/or doing mechanics) to ensure that they don’t cause a wipe, and then doing as much damage as possible. When to hold CDs, what talents to take, etc. could all be traced directly back to those two objectives.
Now, what about healing?
Keep the raid alive
1.1. don’t go oom, so you can keep healing
1.2. cast spells efficiently, so you don’t oom early
Do mechanics
Do as much damage as possible
3.1. cast heals efficiently, so you have more globals to DPS
Optimizing healing is complicated because at the raid level, you’re trying to optimize an entire team of healers to each contribute the right combination of healing, utility and dps to meet the three objectives above. You could also discuss optimizing healing at the individual level, either in the context of contributing to an optimized raid team or in the context of maximizing your own personal contribution. They aren’t always mutually exclusive - but neither do they always go hand in hand.
I’m going to leave disc out of the discussion for now because disc is fairly unique in how it commits to ramps and periods of healing output, and its interaction with dps.
Suppose you have an encounter and raid whose gear/player skill/etc. is tuned to require “50% healing efficiency” from 4 healers. (this is an entirely arbitrary number, and there is no formal definition for “healing efficiency”.) Let’s also assume that there aren’t any particularly unique damage patterns suited to certain healer specs over others and that all healing CDs are interchangeable. (Or just assume all 4 healers are the same spec, whatever works for you.) In other words, assume all variables are constant except for the percentage of time each healer chooses to spend on healing vs. doing dps.
To optimize this particular team of healers, you could have any combination of healing and damage breakdowns so long as it averaged out to 50% healing time across all 4 healers.
Some possible combinations:
2 healers healing 100% of the time and 2 healers dpsing 100% of the time
3 healers healing 66% of the time/dpsing 33% of the time and 1 healer dpsing 100% of the time/only popping their raid CD when assigned (p.s.: this is the setup my raid has had for the last couple tiers,)
3 healers, and drop the 4th healer for an actual dps
etc.
To state the obvious: “healing efficiency”/output is NOT linear. That is to say, assuming that overheal is constant, spending all your time healing will not result in double the healing compared to if you spent half your time healing. Similarly, because CDs are a thing, using your hardest hitting spells on CD is more impactful than spending the same amount of time dpsing but only pressing your filler. I choose to use the concept of healing efficiency as opposed to raw healing numbers because 4 healers healing efficiently while weaving dps is going to result in more net dps, for the CD/non-linear dps increase vs. time reasons mentioned above. I describe “healing efficiency” as if it’s linear, which is incorrect, but it doesn’t change the point of the hypothetical. If this still doesn’t make sense or you think this premise is invalid, then feel free to stop reading now.
Still here? You’ve been warned!
Optimizing healing at the raid level is hard because you want every healer to do just enough healing at the right times to keep the raid nice and healthy, and then dump everything else into dps. In practice, “perfect is the enemy of good [enough].” Overheal is going to happen, healers are going to target the same person with a heal, the healers don’t read each others’ minds, etc etc. It’s better to accept those losses than to ‘penny pinch’ and end up failing the basic healer objective of keeping people alive.
My point is, trying to fully optimize healing and healer dps at the raid level, or even discuss it, is pretty ridiculous because you’re only ever going to be able to take the blunt approach of assigning healer CDs and hoping everyone intuitively scales their output appropriately. Expecting the finesse of not all healing the same target, or "taking turns healing” when each healer’s most efficient spell is off CD, is patently unrealistic.
How about optimizing on the individual level? I describe it simply as “seeking value.” The individual can affect raid optimization in a couple ways.
most obviously, raid CDs are a major efficiency bump that should be coordinated with others.
you can positively affect the mana usage of the other healers (especially reactive healers) by contributing healing when their efficient spells are on CD and/or otherwise “making it so that they don’t need to cast a second, more inefficient spell.”
That last sentence is important. Recall that by my reckoning, optimizing as a healer is not about maximizing hps or drps - it’s about contributing to the ultimate goal of a boss kill by keeping the raid alive, optimizing time/gcd usage, and doing dps.
If a spell would make the difference between life and death, that is the spell you need to use. It is valuable, regardless of ‘efficiency’.
If a spell makes it so other healers don’t need to and do not spend as many globals on healing (and therefore don’t go oom and/or are able to spend more time dpsing), then it is efficient and valuable. (important: in practice, co-healers may cast additional healing spells anyways - maybe to parse, maybe because we’re not mind readers, or any other number of reasons. Even if a spell is efficient on paper, if it hasn’t caused an efficient effect in practice, it has not created value.)
If none of the above apply then maximizing personal contribution simply means healing efficiently/minimizing overheal and doing dps, not much else to say there.
Barrier is interesting because it could fall into either or neither of the above conditions. Examples of the life or death application include healer checks such as siegebreaker+high stacks on Jaina, reckoning on Champions, high energy tantrums on Grong, Blockade if an Ire soak is going to be missed. Examples of the efficient value application include fire/mist on Jadefire, death realm healing on Rasta (could also be life/death if people fail to clear stacks and get targeted by the healing absorb), reactively on Mekka if a bomb clips the entire raid, or on Blockade when lightning is going to hit the ranged stack.
As I have alluded to in previous posts on Evangelism/LB, Shadow Covenant, and parsing, just because something results in high hps according to logs doesn’t mean it’s ‘only’ a pad talent. Furthermore, it doesn’t mean that the talent is morally wrong or that no true mythic raider would take that talent. It’s certainly true that some people may take those talents for the sake of parsing, but to generalize all usages of the spell as selfish padding is bad logic.
Attempt to define your objectives and, when possible, use math and actual logs to determine what is optimal. When dealing with how to cooperatively heal with your co-healers, a healthy dose of intuition could be more useful than getting mired in trying to analyze an overly complex, abstract and non-generalizable healing environment.
